Re: AuthConfReq: Presentational Markup

Karl Dubost, Sat, 27 Mar 2010 18:43:47 -0400:
> Sam, Leif,
> 
> Le 27 mars 2010 à 13:02, Leif Halvard Silli a écrit :
>> Karl: I hope you eventually will confirm that you stand by that claim 
>> ... ;-)
> 
> yes, explanation below.
> 
>> The typographic feature "line-through" can either be "just" that. Or it 
>> can be a signal which tells a story about the edition history that a 
>> particular text has gone through. Or it can be something "in between". 
> 
> Yes but that statement is true for many elements.

The "in between"? Yes.

> Sam's original 
> question was not clear enough and slightly not related to 
> accessibility and even semantics. I could do exactly the same kind of 
> statement with the alt attribute.
> 
> I can choose to put alt="" on a photograph because from my publisher 
> point of view, I think it is just decorative. I'm just making an 
> editorial choice like the choice of a word.

<b> is allowed per HTML5. Should it only permit <strong>? 

> Then Sam tried to clarify by giving options
> http://www.w3.org/mid/4BAE19C5.4070503@intertwingly.net

> 
> * The Conformance Requirements for authors is incomplete
> * The claims are not justified or hit the wrong target

As told in my previous message to this list, <strike> is fairly well 
supported by a group of user agents which the accessibility have used a 
little bit: text based browsers - a tiny bit better supported than 
<del>.

I also think that it is possible to transmit, in a screenreader, that 
the text is striked over. 

> Le 27 mars 2010 à 10:44, Sam Ruby a écrit :
>> As near as I can tell, the use of the <strike> element was not by 
>> Mark himself, but was allowed through whatever gauntlet of 
>> sanitization filters that he employed at the time.
> 
> I relate that to a design flaw of many CMS. Basically it is quite 
> easy to deliver a site which is valid, accessible, SEO friendly, etc. 
> Very often, there are no tools for maintaining the quality. CMS needs 
> a feedback loop. When I publish, I either sanitize (done most of the 
> time for security, not for markup quality), or I give feedback to the 
> person that the written markup is not correct.
> 
> One size doesn't fit all. Some users need a markup validity check 
> feedback, some users are just not tech savy and in this case, the 
> system *should* take care of it. Unfortunately for me, here, I used 
> should. :)

Should the commenter in this case not have been allowed put a strike 
over the text?
-- 
leif halvard silli

Received on Saturday, 27 March 2010 23:38:03 UTC